Social partnership in Greece : is there a Europeanization effect?

Dimitris Tsarouhas

Social Partnership in Greece: Is There a

Europeanization Effect?

ABSTRACT ■How should one understand the influence of Europeanization on social partnership? This article examines the impact of the European Employment Strategy (EES) on Greek social partnership and the role of employers and unions in the formulation of labour market policy. It identifies two potential levels of influence: first-level change which leads to an alteration of policy discourse, and second-level change which opens up space for reforms in policy actors’ preferences and institutional resources. Empirical findings show a loose and indirect link between the advancement of Greek social partnership and the EES. Policy legacies and institutional inadequacies are decisive. Nonetheless, contrasting the Greek and Italian cases reveals the room for manoeuvre available to unions willing to invest in organizational restructuring and a bigger say in the policy process.

KEYWORDS: Europeanization ■Greece ■institutions ■policy change ■social partnership

Introduction

The launch of the European Employment Strategy (EES) was welcomed by some as an innovative way to deal with Europe’s unemployment and welfare state crisis. With time, its agenda expanded to include the goal of strengthening the role of the social partners in policy implementation. To examine the extent to which the EES has altered the structures of social partnership, this article examines developments in Greece both before and after the inauguration of the EES. Greece is a representative case of a Southern European country characterized by weak labour market insti-tutions and low levels of policy concertation. If the process of Europeanization through the EES has strengthened Greek social part-nership despite past deficiencies then this demonstrates a transformative potential; otherwise its influence must be viewed sceptically.

© 2008 SAGE (Los Angeles, London, New Delhi and Singapore)

Introducing a comparative element can enrich our understanding of the factors at work in the evolution of social partnership. I therefore also consider developments in Italy, with which Greece shares a history of adversarial industrial relations, trade union fragmentation and low insti-tutionalization of bargaining procedures (Zambarloukou, 2006). However, in the 1990s Italy became an ‘emblematic case’ of successful social pacts and the rise of policy concertation (Regini, 2000: 16). A com-parative analysis can shed some light on the conditions necessary for transcending adverse institutional legacies.

The article begins with a discussion of the EES, whose aims and objec-tives are conditioned on high levels of social partnership. I set this policy initiative in the wider context of Europeanization, focusing on the expected effects of the Open Method of Coordination (OMC) on na-tional social partnership structures. I then assess the extent to which the EES has produced domestic changes in Greek industrial relations, focus-ing on the conduct of the social partners and the institutional influences of EES on the role of employers and unions. Empirically, the results point to the limited impact of EES and its failure to ‘Europeanize’ Greek industrial relations, and I identify the institutions and actors preventing Europeanization. Theoretically, I stress the salience of mediating vari-ables, such as national legacies and administrative abilities; this underlines the importance of ‘reading’ Europeanization through diverse lenses and on the basis of the political and institutional capacity of actors to take policy initiatives and reverse policy legacies. The Italian trajectory re-inforces the salience of policy legacies but also points to available options on the part of unions to strengthen their role in the policy process. On a macro-level, the article points to the inability of the EES to contribute to the strengthening of social partnership at national level, because of the absence of concrete incentive structures and its overt reliance on national institutional settings to promote closer collaboration.

The European Employment Strategy and OMC

The principal aim of the EES is to promote policy coordination with each member state, working through its own institutional methods towards achieving commonly agreed goals; but underpinning most proposals and policies is an attempt to strengthen the role of the social partners. The lat-ter are viewed as participants in the process, playing an active role in the realization of the adaptability pillar and making direct contributions to National Action Plans (NAPs). ‘No clearer encouragement to national social pactism could be imagined’ (Martin and Ross, 2004: 324).

The European Employment Strategy took shape in the mid-1990s in the context of high unemployment and rising pressures on social expenditure

budgets, and was elaborated at the 1997 Amsterdam Summit as to the con-tent of an Employment Strategy. An Employment Committee was set up to draft Guidelines and monitor progress. The Luxembourg Summit in November 1997 adopted 19 Employment Guidelines centred on adapt-ability (new forms of flexibility), employadapt-ability (emphasis on active labour market policies), equal opportunities, and entrepreneurship (SMEs, start-ups and entrepreneurial skills) (Martin and Ross, 2004). Every member state draws up a NAP explaining how it intends to implement these Guidelines. The Council, the Commission and the Member States scrutin-ize the success of the strategy, while the Commission and the Labour and Social Affairs Council synthesize the National Reports and assess both nation-specific and EU-wide performance. This results in a Joint Employment Report by the Commission and the Council, alongside the Commission’s own Annual Report on employment performance.

Employment Guidelines have been revised annually and extended in scope. After the 2001 Stockholm Summit, the EES has also encompassed horizontal objectives on achieving full employment, promoting quality of work and life-long learning, further incorporating the social partners in the process and tar-geting particular employment rates for different occupational categories. Major changes to the Guidelines were implemented in 2003 (Watt, 2004); following a radical reform in 2005, they are now presented in conjunction with macroeconomic and microeconomic policy guidelines for a three-year period. The so-called ‘Integrated Guidelines for Jobs and Growth’ form the basis for the successors of the NAPs, the National Reform Programmes.

Ashiagbor (2005) sees in the EES an ongoing, unresolved battle between its deregulatory nature stressing flexibility and adaptability and its portrayal of social protection as a productive asset for the European Social Model. Hyman (2001) on the other hand underlines the increasing salience of market forces in determining employment regimes; the ‘moral economy’ counts for little as market-governed norms and assumptions on productivity and competitiveness prevail. Reinforcing this argument, Seferiades (2003) has argued that the attempt to portray the EES as a policy-neutral method distinguished by its consensual char-acter cannot disguise its essentially political nature; like Hyman, he argues that the EES confirms the ‘recommodifying’ character of the employment relationship. The use of the market criterion to deter-mine the suitability of policy choices verifies that it is distinguished by ‘processual novelty’ (Seferiades, 2003: 194) but not much more. Failure to go beyond a market-driven understanding of the employment rela-tionship lies therefore at the heart of criticism of the EES.

The EES involves the most elaborate application of the ‘open method of coordination’ (OMC), a system of intergovernmental cooperation combined with supranational elements offering a compromise between alternative visions of European integration (Jacobsson, 2004: 357). It has

four main elements. First, fixing EU Guidelines and setting timetables for their implementation; second, translating these Guidelines into policy ini-tiatives through specific targets but in a manner sensitive to national and regional differences; third, establishing quantitative and qualitative bench-marks to assess best practice; and fourth, periodic monitoring, evaluation and peer review of the process to facilitate learning (Zeitlin, 2005). The impact of OMC on EU integration is open to diverse interpretations. The most sympathetic accounts treat it as a satisfactory compromise between EU-wide convergence and respect for national diversity (Goetschy, 2003). The OMC is also credited with a capacity to reconfigure policy networks, enhance knowledge-sharing and facilitate innovative solutions to complex issues. Most of all, it is said to have raised the salience of employment and social policy in the EU and at national level. ‘Flexicurity’, advocated as a preventative approach to unemployment, and the importance of ‘acti-vation’, have in many Member States been placed at the core of the politi-cal agenda as a result of the OMC (Barbier, 2005).

Sceptics, however, consider it an attempt to move away from uniform rules on employment and social protection secured through hard law. This could in the long run undermine the European social model. Its non-binding char-acter makes it difficult to correlate, let alone establish causal connections, between national performance and OMC methods as an instrument for pol-icy change. Its effectiveness as a polpol-icy instrument is therefore in doubt. Furthermore, far from establishing a platform for innovative solutions, the OMC can be seen as another form of multilateral surveillance also practised by the OECD and the IMF (Schäfer, 2006). What is more, its reliance on transparency and accountability to promote learning and feedback has often been compromised at the national level. Public and media interest in the process remains low and the design of NAPs has often been interpreted by national ministries in accordance with their own priorities. The involvement of stakeholders, not least the social partners, has thus suffered from an inter-pretation of the OMC emphasizing a traditional intergovernmentalist char-acter (de la Porte and Pochet, 2005). Finally, OMC entails an emphasis on managerial practices similar to the UK paradigm prevalent since the 1980s, privileging goals through target-setting instead of the means requiring the establishment of social dialogue processes (Gold et al., 2007). In that sense, the OMC may sacrifice the institutional advancement of social dialogue at European level on the altar of managerial expediency.

Europeanization and Social Partnership

Europeanization has often been understood as the change or reform of national institutions, policies or indeed polities and their interaction with developments at the EU level. The expansion of policy instruments by

EU institutions, going beyond ‘hard law’ issues and encompassing ‘soft law’ policy prescriptions, has reignited the debate on the meaning of Europeanization. In a definition that captures the essence of its multi-dimensional character, Radaelli sees it as consisting of processes of con-struction, diffusion and institutionalization. These include paradigms, beliefs and ‘ways of doing things’ that have been first defined at EU level and subsequently absorbed by governments and regions on the level of public discourse and policy (Radaelli, 2003). Europeanization has differ-ent meanings depending on the policy area under investigation, and need not only involve the top-down impact of the Union on Member States (integration). It is necessary to understand domestic institutional and organizational interaction before assessing whether Europeanization has played any role in revising or modifying the original domestic policy constellation (Radaelli, 2005). In that sense, it is to be measured not only with reference to EU actions, but also by the way domestic actors refer to European institutions and policy-making processes.

Social partnership at national level is an instructive issue to explore when studying Europeanization in accordance with this approach. Europeanization involves the process of learning from others and the potential imitation of successful policy paradigms. The EES has been described as a forum for the dissemination of policy learning, promoting ideational convergence on acceptable and unacceptable policies, as well as the assessment of mechanisms most likely to produce the desired results. Benchmarking in particular can be utilized to promote and export ‘best practice’ from one country to another, including in the field of social partnership (Vos, 2006). The use of benchmarks as learning instruments can alter the way policy-making is understood and practised (De la Porte and Pochet, 2004). Its iterative, cross-cutting and delib-erative nature facilitates the exchange of information and the engagement of diverse actors in the process of policy formulation.

Applying a modified version of the classification used by Hall (1993) in his seminal study on policy change (Hall, 1993), two types of impact of the EES can be distinguished. First-level change entails that policy actors modify their language in discussing and analyzing issues, perceive prob-lems and reform points in a European rather than national context, draw comparisons with their EU counterparts regarding policy reform and widen their agenda to address EES requirements. Second-level change means that social partners become actively involved in labour market pol-icy formulation and cooperate closely to implement policies, triggering institutional change through a change in their sets of preferences.

Europeanization in Greece

I treat Europeanization as a mediating factor in explaining change in Greek social partnership. Greece is a Southern European country seeking

to Europeanize its public policy structures by modernizing its adminis-trative and organizational apparatus. During the 1990s, a general consen-sus emerged among policy-makers that convergence with EU standards ought to be a priority so that Greece could join EMU as soon as possible. ‘Convergence with Europe’ articulated both a pragmatic policy goal, safe-guarding the drachma from market volatility, but was also part of the modernization process aiming at the acceptance of Greece by the EU as an equal partner. Policies and ideas are interwoven, the latter reinforcing the drive towards the adoption of policies conducive to the realization of Europeanization understood as modernization.

Using a slightly modified categorization from the one developed by Jacobsson (2004), I outline a set of practices that, though not unique to OMC, are systematically used by it to promote policy change. These are: a) the changed use of language towards a Europeanized framework of action, b) developing common statistical indicators and other scientific tools, c) the strategic use of comparisons and evaluation with European counterparts, and d) knowledge diffusion and the broadening of the social partners’ policy agenda. Relying on the identification of such mechanisms to detect first- and second-level changes, I apply them to the case of the Greek social partners to measure the impact of the EES. The hypothesis is that because the EES relies on national administrative and organizational structures it is too weak to trigger processes of domestic policy transformation regarding the operation of social partnership. In such conditions, the role of mediating factors in explaining policy out-comes beout-comes all the more important.

Greek Social Partnership

To assess the impact of the EES on national settings, the point of depart-ure needs to be the domestic context. Problems, resources and ideas prevalent at that level should be analysed before we attempt to deduce the empirical and/or transformative influence of a strategy based on OMC.

In Greece, the main social partners on the union side are the General Confederation of Labour (GSEE) and the Confederation of Public Servants (ADEDY), and on the employers’ side the Federation of Greek Industry (SEV), the National Confederation of Commerce (ESEE) and the General Confederation of Small Businesses and Trades (GSEVEE). Their relationship with the state has traditionally been problematic. Excessive reliance on party links, the domination of civil society by politi-cal parties, and networks of clientelism contributed to the formation of ‘disjointed corporatism’ (Lavdas, 1997) as the main form of interest rep-resentation. A plethora of associations and groups divided employers and employees into interest groups lacking coherence and strong organiza-tional resources, preventing their emancipation from state structures.

Greece’s turbulent historical past and weak administrative system have rendered the process of Europeanization one-dimensional, at least with regard to the transfer of policy competences. While the EES obscures the boundaries of policy transfer from Brussels to Member States or vice versa because of its non-binding character and the mix of EU and national organs, Greece is clearly a follower rather than a leader.

Before democratization in 1974, the industrial relations system was characterized by restrictive labour practices and the exclusion of inde-pendent trade unions. Collective bargaining was state sponsored and sanctioned under the terms of the law of 1955, leading to a centralized and hierarchical structure. It mandated the reference of differences to arbitration tribunals, providing limited possibilities for collective bar-gaining. Attempts to institutionalize structures conducive to social dia-logue, such as the Social Policy Council and the National Development and Productivity Council, failed.

In the 1980s, a series of changes led to the transformation of Greek social partnership and the emergence of concerted action on the part of labour and business representatives. First, the Socialist Party (PASOK), in government for most of the decade, toned down its anti-business rhet-oric after 1985 and SEV acquired a dominant position in business repre-sentation. Its coverage and membership grew rapidly in the late 1980s and early 1990s (Lanza and Lavdas, 2000). Second, the liberalization of the legal framework regarding interest representation assigned equal sta-tus to business and labour as legitimate representatives of their respective members. In 1990 the National Unity government adopted a law with the assent of both GSEE and the three employer organizations, institu-tionalizing free collective bargaining. It also created a framework for decentralized bargaining and abolished compulsory arbitration. For the first time, two new levels were recognized in the bargaining sphere, sec-toral and enterprise, and the traditional craft-based representation was made irrelevant. The two new tiers dominate collective bargaining and complement the centralized National General Collective Labour Agreement (EGSEE) signed between SEV and GSEE every few years since the early 1990s, which sets minimum wages and basic labour con-ditions, is legally binding and covers all employment relationships. In the new framework, the incentives for commonly agreed solutions have been greatly enhanced (Zambarloukou, 2006), and central agreements have so far been reached without the need for arbitration. The last EGSEE was signed in spring 2008 for a two-year period. Finally, a law in 1999 intro-duced collective bargaining to the public sector.

Third, new institutions were created with the explicit aim of promoting social dialogue. The most important are the Arbitration and Mediation Organization (OMED), the Economic and Social Committee (OKE) – which must be consulted over industrial relations legislation and can also

issue opinions on its own initiative on general issues of economic and social policy – and the National Employment Council (NEC). Moreover, the social partners agreed in 1994 to create a Fund for Employment and Vocational Training (LAEK), which they jointly finance and administer and which received statutory backing. A similar process of institutional innovation based on general collective agreement led to the creation of the Hellenic Institute for Occupational Health and Safety (ELINYAE) in 1993.

The Impact of EES on Greek Social Partnership

First-Level Change

The EES relies to a great extent on social mechanisms to effect the for-mulation and execution of policy. There is evidence of a changing dis-course on the part of the Greek social partners in the 1990s. Instead of the EES, however, it is the institution-building process that seems to be the key mechanism for this change.

OKE is an example: its 2002 opinion on the progress of Social Dialogue emphasized that ‘the social dialogue in all its forms is a component part of the European Social Model, which incorporates values such as responsi-bility, solidarity, participation and joint action’ (OKE, 2002). A second example is LAEK and ELINYAE. Although EU activity in the field of vocational training through CEDEFOP may have provided the platform for final agreement on LAEK, the fact that the agreement was signed as part of the EGSEE points to the salience of the domestic institutional structure created after 1990. The creation of the Labour Institute (INE) by GSEE and ADEDY indicates their willingness to analyse labour mar-ket issues in a systematic and scientific manner. Evidence suggests that the INE is engaged in regular cross-European comparisons to determine income, taxation and social benefit levels in the EU.

Nevertheless, trade unions continue to operate in a more traditional framework when it comes to labour and economic policy. The Information Newsletters released by the INE between 1995 and 2003 cover 24 thematic categories, only one of which relates to ‘EU-EMU-Social Europe’ in a systematic manner. While the number of contribu-tions and attempts to disseminate information in this category is large (56 articles in total), the distinction between the domestic and European sphere is clear. The Greek social partners have traditionally approached the country’s labour market problems from an interest group perspective concerned with everyday issues and ignoring the wider socio-economic context of their demands. This has had a negative impact on their ability to assume responsibility for policy reform.

While the inauguration of the EES in 1997 hardly affected the policy orientation of the social partners, the Lisbon Agenda influenced their discourse on economic reform. Agreeing on the goals of more competi-tiveness and ‘flexicurity’ meant that both employers and unions acquired a new point of reference in formulating their strategies. SEV has created a separate section in its Positions’ Archive entitled ‘The Lisbon Strategy’. The unions’ reports on Greek economic policy are increasingly formu-lated in a European context, drawing parallels with developments in Spain, Italy and Portugal. Annual Reports published by the INE since 1999 are indicative: in their 2005 report spanning the first half of the Lisbon process, they called for the creation of a ‘knowledge economy’, emphasizing employee retraining and the enhancement of knowledge-intensive activities to offset low-wage pressures from neighbouring Balkan countries (EIROnline, 2006). A Europeanization of the social partners’ rhetoric is increasingly discernible, adding an explicitly EU layer prior to the Greek one on issues pertaining to economic and labour market policy reform.

At an empirical level, the impact of the EES on Greek social partnership is limited. The liberalization of the legislative framework on employment in the 1990s is more important in shifting the direction of policy behav-iour. State initiatives were decisive in strengthening the legislative frame-work. Examples are the creation of the National Employment Council and the National Committee for Social Exclusion following the 2002 Employment Guidelines. This is not to dismiss the salience of the ‘Europe’ factor. The Lisbon Agenda has intensified the use of the European yardstick regarding policy proposals, and research analysis by both sides is increasingly conceptualized in a European framework. This may be due to the concrete goals established after the Lisbon and Stockholm Councils, which allowed governments and social partners to utilize a concrete framework to achieve their aims. Still, it is obvious that convergence in ‘talk’ does not mean convergence in decision-making. The ‘Lisbon’ language is now used, but the radically different meanings attri-buted to flexibility by employers and unions, and the consequences of this divergence regarding social partnership, remain unchanged.

Second-Level Change

The second set of variables employed to study the impact of the EES encompasses the degree of change of the social partners’ agenda, and the strengthening of their role through their participation and implementa-tion role in NAP. This is easier to measure compared to empirical change because data are more accessible. It is also more important in terms of policy outcomes, since a change in discourse may be more con-fusing than revealing. ‘The institutionalization of ideas is a problematic

process … Communities of discourse may produce symbolic policies’ (Radaelli, 2003).

Strengthening the role of employers and unions has often been linked to the reinvigoration of social pactism and policy concertation in the 1990s, albeit in conditions of a dominant supply-side agenda and overt emphasis on the link between social pacts, enhanced competitiveness and structural economic reforms (Ferner and Hyman, 1998). They have been reached on a bipartite (Finland, Netherlands) or tripartite (Ireland, Italy and Portugal) basis. In countries with no previous experience in central-ized bargaining and low levels of concertation such pacts have implied a regime shift (Vos, 2006). This was nowhere more evident than in Italy, where the increasing use of social pacts over the 1990s contributed to the (temporary) restoration of competitiveness through trilateral agreements on incomes policies and social security reform (Zambarloukou, 2006). In Greece, attempts to emulate such agreements started in 1997. In an effort to adapt the Greek economy to EMU requirements, the government invited the social partners to talks aiming at the institutionalization of social dialogue procedures. In November, a Confidence Pact ‘towards the year 2000’ was signed. However, its content was vague and no agree-ment was reached as to the desired direction of reform. The process was repeated in 2000 and 2001 to reform the labour market and social securi-ty system in a way similar to processes already concluded or under way elsewhere in the EU. Both employers and unions declared their dissatis-faction, as neither side managed to persuade the government to introduce its preferred type of legislation.

For all the air of disappointment that covered these deliberations, evi-dence suggests that employers and unions have made attempts to act in accordance with a ‘more “grown up”, socially responsible profile’ (Papadimitriou, 2005: 392). Though it is impossible to determine whether this is a direct effect of socialization in EU circles, similar negotiations in other member states and the pressures for reform emanating from the EU are likely to have played a role. There are indications that the social part-ners are dropping their old dependence on the state to reach agreements. In 1999, GSEE and SEV established a Joint Committee to study reduc-tions in working time, following the example of EU states where such a reform had materialized. While the Committee’s deliberations ended in disagreement, the two sides sought to reach a bilateral agreement on labour reform a year later, urging the government not to legislate before the end of their talks. The latter, however, went ahead with its own reform that clashed with both sides’ priorities (Papadimitriou, 2005).

The involvement of the social partners in NAP had been rudimentary and of little value until 2001. In that year, the social partners were for the first time able to participate in the drafting of the NAP and the Third Community Support Framework. Even then, their views were merely

considered in the final NAP, with no suggested input incorporated in the final text. In both the planning and implementation stage, the social part-ners were essentially left out of the process and complained of insuffi-cient consultation (EIROnline, 2004). Moreover, there has been no attempt to synthesize the two sides’ views and propose common actions, as differences remain too large to bridge.

Another channel of potential influence and coordination by employers and unions has been the creation of Local Employment Pacts (TSAs). These began as a Commission initiative to boost local employment. In the Greek context they were meant to translate into special collective agreements at local level. Vagueness characterized who was to participate and which areas the agreements were meant to cover. Unions saw them as an attempt to deregulate the labour market while employers derided the ‘statist’ character of their set-up. In the end, the absence of incentives for employers’ participation and SEV demands that collective agreement provisions on wage increases be not applied in TSA areas meant that the experiment failed (OKE, 2002). Although TSAs received statutory back-ing in 1998, they have never been put into practice.

The Limits of the EES and the Role of Domestic Factors

The impact of Europeanization on the Greek social partners has to date been merely indirect, and inadequate to assist the strengthening of their role. While there is evidence of ‘empirical Europeanization’, albeit in conditions and ways only loosely related to the EES, the policy-oriented impact of the EES has been very limited. The social partners’ role in the formulation and implementation of NAPs has been rudimentary and there have been no attempts to formulate common positions. What needs to be analysed, therefore, is the salience of mediating variables related to domestic institutional and political conditions in transmitting policy ideas from the European to the national level, and the internalization of policy norms stemming from the EU. A useful starting point is the set of characteristics of the social partners and the state prior to the onset of Europeanization, that is, the historical evolution and institutional struc-tures shaping the policy outlook of the two sides prior to EU member-ship. ‘The existence of norm entrepreneurs and consensus-oriented cultures affect whether European ideas, norms and the collective under-standings which do not resonate at the domestic level, are internalized by domestic actors giving rise to domestic change’ (Börzel and Risse, 2000: 9). The absence of such characteristics is likely to impede change. Both of these characteristics are present in the Greek case.

On the one hand, traditionally conflict-ridden industrial relations are linked to the intense politicization of trade unions and state control over

their activities. For the greatest part of trade union history, their legiti-macy depended on state control. In the 1920s the practice of appointing state-sponsored union leaderships became widespread, as did the state financing of workers’ organizations. The formation of the Labour Home (EE) in 1931 is an example. Operating under the auspices of the Finance Ministry, the EE manages the funds generated from the contri-butions made by employers and employees, prolonging the dependency of unions on state funding (Lavdas, 2005).

This results in a weak trade union movement. Its ability to articulate an agenda distinct from the state in the post-1974 era was replaced by party political patronage. The PASOK government of the early 1980s criticized the authoritarian and clientelistic state of the past, attempting to shift the political balance in its favour. Personalized clientelism was replaced by clientelism channelled through the party and the state machinery. The clearest articulation of this practice was the imposition of proportional rep-resentation in the elections of the GSEE presidium. State control of union activity was replaced by party political control. In 1987 the government proceeded with legislation on the ‘democratization of professional associ-ations’, according to which proportional representation was established in the elections of GSEEVE too. Meanwhile a 1982 law forbids the formation of trade unions in firms employing fewer than 20 employees, which given the size and structure of Greek business excludes unions from 99.5 percent of private enterprises. Policy legacies stemming from the mode of inter-action between the state and the unions hampers the ability of the latter to acquire an autonomous societal and eventually political role.

The frequent description of the post-1974 Greek reality as state cor-poratist misses an essential point. State attempts to control the unions went hand in hand with its inability to impose its own developmental agenda. State–union relations were integrated in a wider network of clientelistic practices, where pluralistic societal demands were put on the agenda of an ineffective and malfunctioning state apparatus (Pagoulatos, 2003). It is for such reasons that the term ‘disjointed corporatism’ (Lavdas, 1997) is more suitable to the Greek experience. Strong state links apply to employers too, though the post-1974 picture is again mixed. On the one hand there were increasing pluralist tendencies and emancipatory moves away from state dependence, and on the other per-sistent sectoral corporatist practices in areas such as shipping (Lanza and Lavdas, 2000). Overall, employers have been in close collaboration with successive governments to promote their agenda. Placing the 1990s legis-lative reforms in their historical context provides important insights as to the social partners’ inability to acquire a prominent role in the EES process and Europeanize their modus operandi.

A second key factor behind the inability of the social partners to gain a meaningful role in public policy is their fragmented nature. Although their

organizational structure appears unitary at the peak level, divergent fac-tions are in existence within the GSEE (DAKE, PASKE, ESAK, PAME) corresponding to party political groupings. Fragmentation at the peak level is accompanied by a plethora of primary level unions, estimated at about 7000. These maintain operational independence from the second tier of organization, namely the 100 or so federations and labour centres (Lavdas, 2005). Changing social partnership is made more difficult if one considers that interest groups have ‘historically been considered [by the state] to be particularistic, and … disruptive factors’ (Lavdas, 2005: 298). On the employers’ side, SEV has historically maintained a higher degree of auton-omy from the state to promote a clear line of interest representation. On the other hand, its membership encompasses both individual firms and associations. While it records high levels of the latter’s membership, it does less well in attracting firms and regional associations. Business interests are still organized on a sectoral level and since the 1950s SEV’s role as a peak confederation has been limited in its success.

Greek political culture has traditionally been characterized by the lack of broad-based consensus and conciliatory mechanisms to facilitate reform. The reasons for such an absence need to be traced back to the his-torical evolution of the Greek polity distinguished by major cleavages along political lines, economic instability, and political radicalism. Civil war during the 20th century and a lack of accomodationist political culture deprived the country’s social partners from a critically important platform on which they could base their partnership once formal institutional changes had been introduced. Institutions matter, and their recent reform has had some positive impact on Greek industrial relations. Radical shift cannot, however, be accomplished in the face of overwhelming policy leg-acies and persistent institutional inadequleg-acies that have domestic origins but are also linked to the failure of the EES to follow through its rhetori-cal commitment to social partnership.

Changes introduced in the 1980s on the democratization of interest representation proved superficial. Social partnership retained a rubber-stamping character. Decisions already taken at the state level were merely communicated to the social partners. Changes in social partnership have not been the result of establishing norm entrepreneurs, or the overcoming of fragmentation in the employers’ and employees’ representative organs. Change has been the result of political expediencies and political priorities as formulated by different governments. Finally, agreement to move ahead with common propositions on some areas has been the exception to the rule. To this day, institutions such as OKE function on an overtly plural-istic and purely advisory role, registering the divergent views of their members.

Fragmentation and the absence of a consensus-seeking culture are also characteristic of Italy. Over the 1990s, however, social partnership

flourished. In 1993 the Amato government offered unions a more coor-dinated system of collective bargaining in exchange for wage moderation and the abolition of scala mobile (Zambarloukou, 2006). Though the lat-ter had been unilalat-terally repudiated by the employers in 1992, the unions’ participation in the agreement meant that they gained a say in policy-making and helped shape new rules on bargaining (Molina Romo, 2005). In 1995, agreement on pension reform was reached between the technocratic government of Dini and the unions. The employers’ with-drawal from the process allowed the unions to further strengthen their position. More agreements followed in 1996 through the tripartite ‘Jobs Pact’ and in 1998, when the ‘Pact for Development and Employment’ was signed. Both agreements sought to increase labour market flexibility and lighten the employers’ tax burden in return for more investment in education and training (Della Sala, 2004).

Explanations for the strengthening of social partnership in Italy point to domestic factors. They range from the degree of involvement in EMU and the need to cooperate within a crisis-ridden political climate, to the country’s economic structure, and the apparent conversion towards an intermediate model identified with the German industrial relations regime (Regini, 2000). However, all of the above are riddled with incon-sistencies and are not applicable to similar cases elsewhere in Europe. Greece and the Netherlands are a counter-example to the EMU factor; Sweden and Germany to the weakness and convergence theses. A better explanation needs to address contingent political factors and the util-ization of policy legacies by the unions to enhance their role.

The extraordinary conditions of the early 1990s in Italy did not facili-tate partnership per se. In fact, the 1993 agreement was only reached fol-lowing the intervention of the Amato government amidst continuing dis-agreement between the social partners. The collapse of the old political system and the identification of large sections of business with the scan-dals that precipitated its downfall facilitated union attempts to reorgan-ize their operation. The unions utilreorgan-ized the political vacuum and, aware of their own fragmentation, established a bargaining structure to strengthen their rank-and-file representatives and enhance their political role (Molina Romo, 2005). The process started in 1991 when agreement was reached among the three major confederations on representation at company level (Della Sella, 2004). The usefulness of local unionism was made apparent after the successful referenda on the 1995 pension reform revealed the practical benefits of enhancing internal democracy. Institutional design combined with agency mobilization made a differ-ence. A similar process has not occurred in Greece.

A second factor that distinguishes the Italian case is the ability of unions to create widespread acceptance of reforms to enter EMU. The different

policy legacies with Greece are important. While the Italian unions could assume responsibility for reform assisted by a large-scale political crisis, they could also fall back on Italy’s ‘anchoring strategy’ to Europe pursued by all Italian governments after the war. In the 1970s, the Italian unions were opposed to the European Monetary System (EMS) but distinguished it from European integration to which they were positively predisposed. When the Andreotti government decided to join the EMS in 1978, Italy embarked on a long process of soul searching to identify policies that would strengthen its vincolo esterno (Ferrera, 2004). In contrast, the conversion of Greek society in general and the unions in particular to adherence of the EU project had a much shorter historical record. While by the beginning of the 1990s the change from cautious to enthusiastic Europeans was secured, the PASOK record of the 1980s allowed large parts of the union movement to equate reforms ‘imposed by Brussels’ with neoliberalism. The close links between the unions and PASOK in the absence of widespread crisis conditions meant that the costs of accepting reforms in the name of a project until recently portrayed as antithetical to the national and party interest were disproportionately high.

Conclusion

Europeanization of industrial relations in the case of Greek social part-nership means the successful domestic adaptation of the conciliatory social partnership paradigm that flourished through the reinvigoration of social pacts elsewhere in the EU, and especially in Italy, over the 1990s. The EES seeks to reform Europe’s labour markets by way of strengthen-ing the participation of social partners in the process of policy reform. In its five-year report following the inauguration of the EES, however, the Commission acknowledged limited progress regarding the involvement of social partners in policy formulation and implementation (CEC, 2002). Integrating the social partners is meant to make policy reform long-term oriented and minimize the risks of inertia by spreading partic-ipation and dispersing costs. However, the EES suffers from a lack of adequate incentive structures regarding the participation of the social partners in policy design and implementation.

Greece is a case in point. The evidence points to limited change in Greek social partnership stemming from the EES process, a rhetorical shift and a few institutional amendments notwithstanding. The failure to Europeanize beyond a limited extent points to the inadequacy of learning through networks of facilitated policy coordination in the presence of his-torical and institutional factors blocking the re-articulation of policy pref-erences along conciliatory lines. Though a complete overhaul of social

partnership as a result of EES always seemed unlikely, even rudimentary changes are difficult to detect. With regard to social partner participation in the NAPs in particular, our findings confirm a trend observable else-where in Europe. ‘In those countries else-where there is no tradition of partic-ipation by the social agencies in the taking of decisions, there was no opening up to greater participation’ (Garcia et al., 2004).

This is confirmed in the case of Greece. A series of mediating variables disrupted the process of Europeanization of industrial relations despite the institutional reform of the 1990s. The heavy politicization of indus-trial relations until the early 1990s and the fragmented nature of both unions and employers have had a negative effect on the ability of social partners to conclude long-lasting agreements and influence government policy. Attempts by the state to intervene in the social partners’ organi-zational structures and the superiority syndrome of the state apparatus meant that unions and employers have played a marginal role in the for-mulation and implementation of NAPs. On the other hand, initiatives and common positions increasingly adopted over the last ten years are a sign that things may be about to change, signifying a first-level change of approach.

The Italian comparison reveals both the salience of policy legacies but also the institutional room for manoeuvre on the part of the unions. In Italy, the proliferation of social pacts in the 1990s and the increase in concertation were a function of contingent political factors, shrewd insti-tutional manoeuvring by the unions and a favourable starting point regarding the historical importance of integration with Europe. While policy legacies can only be reversed with time, possibilities exist for Greek unions to strengthen their organizational abilities at the local level, relate employees to their activities and enhance their political status as a result. Concretely, the presence of norm entrepreneurs ready to overhaul the deficient organizational framework of trade unionism could make a difference to the unions’ salience in public policy and the coherence of their stance on policy reform. This is not to say that conflicting positions and priorities would disappear: the Italian case is instructive as to the heightened conflicts between the social partners on labour market flexi-bility after EMU entry. It also points, however, to the possiflexi-bility of enhancing the unions’ political role through the renewal of their organi-zational structure.

ACKNOWLEDGEMENTS

An earlier version of this article was presented at the annual conference of the Political Studies Association, University of Reading, April 2006. I thank George Pagoulatos, the editor of EJIR and two anonymous referees for their helpful comments and suggestions.

REFERENCES

Ashiagbor, D. (2005) The European Employment Strategy: Labour Market Regulation and New Governance. Oxford: Oxford University Press. Barbier, J.-C. (2005) ‘The European Employment Strategy: A Channel for

Activating Social Protection?’, in J. Zeitlin, P. Pochet and L. Magnusson (eds) The Open Method of Coordination in Action: The European Employment and Social Inclusion Strategies, pp. 417–46. Brussels: P.I.E. Peter Lang. Börzel, T. and Risse, T. (2000) ‘When Europe Hits Home: Europeanization and

Domestic Change’, European Integration Online Papers 4(15), available online at: [http://eiop.or.at/eiop/texte/2000–015a.htm].

CEC (2002) ‘Taking Stock of Five Years of the European Employment Strategy’, COM (2002) 416, CEC, Brussels.

De la Porte, C. and Pochet, P. (2004) ‘The European Employment Strategy: existing research and remaining questions’, Journal of European Social Policy 14(1): 71–8.

De la Porte, D. and Pochet, P. (2005) ‘Participation in the Open Method of Coordination: The Cases of Employment and Social Inclusion’, in J. Zeitlin, P. Pochet and L. Magnusson (eds) The Open Method of Coordination in Action: The European Employment and Social Inclusion Strategies, pp. 353–90. Brussels: P.I.E. Peter Lang.

Della Sala, V. (2004) ‘Maastricht to Modernization: EMU and the Italian Social State’, in A. Martin and G. Ross (eds) Euros and Europeans: Monetary Integration and the European Model of Society, pp. 126–49. Cambridge: Cambridge University Press.

EIROnline (2004) ‘Social Partner Involvement in the 2002/3 National Action Plans’, available online at: [http://www.eurofound.eu.int/publications/files/ EF04144EN.pdf.].

EIROnline (2006) ‘Social Partners Disagree at National Employment Council’, available online at: [http://www.eiro.eurofound.ie/2006/02/inbrief/

gr0602102n.html].

Ferner, A. and Hyman, R. (1998) ‘Introduction: Towards European Industrial Relations?’, in A. Ferner and R. Hyman (eds) Changing Industrial Relations in Europe, pp. xi–xxvi. Oxford: Blackwell.

Ferrera, M. (2004) Rescued by Europe? Social and Labour Market Reforms in Italy from Maastricht to Berlusconi. Amsterdam: Amsterdam University Press.

Garcia, M, Salzmann, A.C. and Pradel, M. (2004) ‘The European Employment Strategy: An Example of European Multi-level Governance’, SEI Working Paper No.77.

Goetschy, J. (2003) ‘The Employment Strategy and European Integration’, in D. Foden and L. Magnusson (eds) Five Years’ Experience of the Luxembourg Employment Strategy, pp. 69–109. ETUI: Brussels.

Gold, M., Cressey, P. and Léonard, E. (2007) ‘Whatever Happened to Social Dialogue? From Partnership to Managerialism in the EU Employment Agenda’, European Journal of Industrial Relations 13(1): 7–25.

Hall, P.A. (1993) ‘Policy Paradigms, Social Learning, and the State: The Case of Economic Policymaking in Britain’, Comparative Politics 25(3): 275–96.

Hyman, R. (2001) ‘The Europeanisation – or the Erosion – of Industrial Relations?’, Industrial Relations Journal 32(4): 280–94.

Jacobsson, K. (2004) ‘Soft Regulation and the Subtle Transformation of States: The Case of EU Employment Policy’, Journal of European Social Policy 14(4): 355–70.

Labour Institute (2005) ‘I Elliniki Oikonomia kai h apasxolisi’, available online at: [http://www.inegsee.gr/ekthesi2005/ekthesi2005.htm].

Lanza, O. and Lavdas, K. (2000) ‘The Disentanglement of Interest Politics: Business Associability, the Parties and Policy in Italy and Greece’, European Journal of Political Research 37: 203–35.

Lavdas, K. (1997) The Europeanization of Greece: Interest Politics and the Crises of Integration. Basingstoke: Macmillan.

Lavdas, K. (2005) ‘Interest Groups in Disjointed Corporatism: Social Dialogue in Greece and European “Competitive Corporatism”’, West European Politics 28(2): 297–316.

Martin, A. and Ross, G. (2004) ‘Conclusions’, in A. Martin and G. Ross (eds) Euros and Europeans. Monetary Integration and the European Model of Society, pp. 309–30. Cambridge: Cambridge University Press.

Molina Romo, O. (2005) ‘Political Exchange and Bargaining Reform in Italy and Spain’, European Journal of Industrial Relations 11(1): 7–26.

OKE (2002) Opinion No.86. Social Dialogue in Greece: Evaluation, Trends, Prospects. Athens: OKE.

Pagoulatos, G. (2003) Greece’s New Political Economy: State, Finance and Growth from Postwar to EMU. Basingstoke: Palgrave Macmillan. Papadimitriou, D. (2005) ‘The Limits of Engineering Collective Escape: The

2000 Reform of the Greek Labour Market’, West European Politics 28(2): 381–401.

Radaelli, C. (2003) ‘The Open Method of Coordination: A New Governance Architecture for the European Union?’, Sieps Preliminary Report, EO Print, Stockholm.

Radaelli, C. (2005) ‘Europeanization: “Solution” or “Problem”?’, paper for the APSA Annual Meeting, Washington, DC, September.

Regini, M. (2000) ‘Between Deregulation and Social Pacts: The Responses of European Economies to Globalization’, Politics and Society 28(1): 5–33. Schäfer, A. (2006) ‘A New Form of Governance? Comparing the Open

Method of Co-ordination to Multilateral Surveillance by the IMF and the OECD’, Journal of European Public Policy 13(1): 70–88.

Seferiades, S. (2003) ‘The European Employment Strategy Against a Greek Benchmark: A Critique’, European Journal of Industrial Relations 9(2): 189–203.

Vos, K.J. (2006) ‘Europeanization and Convergence in Industrial Relations’, European Journal of Industrial Relations 12(3): 311–27.

Watt, A. (2004) ‘Reform of the European Employment Strategy after Five Years’, European Journal of Industrial Relations 10(2): 117–37.

Zambarloukou, S. (2006) ‘Collective Bargaining and Social Pacts: Greece in Comparative Perspective’, European Journal of Industrial Relations 12(2): 211–29.

Zeitlin, J. (2005) ‘Introduction: The Open Method of Co-ordination in Question’, in J. Zeitlin, P. Pochet and L. Magnusson (eds) The Open Method of Coordination in Action: The European Employment and Social Inclusion Strategies, pp.11–25. Brussels: PIE Peter Lang.

DIMITRIS TSAROUHAS is Assistant Professor in the Department of International Relations, Bilkent University, Turkey.

ADDRESS: Department of International Relations, Bilkent University, Bilkent 06800, Ankara, Turkey. [e-mail: [email protected]]